Cluster curves based on FPCA scores

Usage

functional_pca_clusters(
  x,
  k = NULL,
  components = NULL,
  method = c("kmeans", "hclust"),
  choose_k = c("none", "silhouette", "elbow"),
  ...
)

Arguments

x: An object of class "r4pde_functional_pca".
k: Number of clusters.
components: Integer vector of components to use.
method: Clustering method: "kmeans" or "hclust".
choose_k: Method used to choose k when k = NULL: "silhouette" (largest mean silhouette width over k = 2 to 10) or "elbow" (largest drop below the line joining the within-cluster sums of squares of k = 1 and k = 10). "none" requires k. The criterion table is attached to the result as attribute "k_selection".
...: Additional arguments passed to clustering functions.